MrNeRF/awesome-3D-gaussian-splatting

Curated list of papers and resources focused on 3D Gaussian Splatting, intended to keep pace with the anticipated surge of research in the coming months.

MrNeRF/awesome-3D-gaussian-splatting is a curated resource list (an "awesome list") for 3D Gaussian Splatting (3DGS) — a technique for rendering 3D scenes from photos or video in real time. The project collects links to papers, code, tools, and learning material, all in one place.

What's inside

The README organizes resources into five main sections:

1. Papers & Documentation

  • A searchable database of 3DGS papers (hosted on the project's website).
  • Courses & tutorials, including an MIT lecture series on inverse rendering and a tutorial by the original 3DGS paper authors.
  • A dataset link (NERDS 360) for outdoor scenes.

2. Implementations

  • The official reference implementation by the original 3DGS authors (graphdeco-inria).
  • A table of community implementations in various languages (C++, Python, Taichi, Warp, etc.) with licenses and short descriptions. Examples include LichtFeld Studio (a native desktop app), Nerfstudio's gsplat, OpenSplat, and several reimplementations.
  • Frameworks that build on or extend 3DGS, such as Pointrix, GauStudio, and DriveStudio.

3. Viewers & Game Engine Support

  • Game engine plugins for Unity, Unreal, PlayCanvas, and Godot.
  • Web viewers split into WebGL and WebGPU categories (e.g., Splat Viewer, gsplat.js, SuperSplat Viewer).
  • Desktop viewers (e.g., 3DGS.cpp, vkgs, splatviz) and native applications like Blender add-ons, a Houdini renderer, an iOS Metal viewer, VR support, and ROS2 support.

4. Tools & Utilities

  • Data processing tools: format converters (PLY, SPZ, SPLAT, etc.), point cloud editors, and pipelines that turn video into 3DGS scenes.
  • Development tools: Houdini integration, camera parameter conversion, and a browser-based 3DGS editor (SuperSplat).

5. Learning Resources

  • Blog posts explaining 3DGS from introductory to technical levels, including compression techniques and comparisons with NeRFs.
  • Talks (e.g., a Metaverse Standards Forum discussion on standardization).
  • Video tutorials and a YouTube channel for LichtFeld Studio.

Key takeaway

This is not software itself — it's a directory of links to software, papers, and learning material. If you want to learn about 3DGS, find code to try, or pick a viewer for your platform, this list is a starting point. The project also invites contributions (via a CONTRIBUTING.md file) and credits several people who helped maintain it.

Related

  • Project
    lightningpixel/modly

    Modly is an open‑source desktop app that runs AI models locally on your GPU to convert a single photo into a 3D mesh. It works on Windows, Linux and Apple‑Silicon macOS, supports NVIDIA, AMD (ROCm) and Apple GPUs, and offers an extensible plugin system for adding new 3D generation models. A graphical workflow editor and a Python CLI let you use it interactively or automate it.

  • Project
    carson-katri/dream-textures

    A Blender add-on that integrates Stable Diffusion to generate seamless textures, concept art, and image assets directly within the 3D workspace.

  • Project
    apple-aiml-research/ml-sharp

    SHARP is a fast monocular view synthesis tool that uses a neural network to generate a metric 3D Gaussian representation from a single image in under a second.

  • Project
    facebookresearch/pytorch3d

    A PyTorch-based library for 3D Computer Vision research providing differentiable rendering and efficient tools for manipulating 3D meshes and point clouds.